Good read, I think Dave and I would get along just great. I’m also firmly in the camp of AI is great and we should all be using it but we need better measured outcome-oriented analysis of it. curlewis.co.nz/posts/lines-o…

Fable is a good model. As with all new models, it is simultaneously excellent and entirely unremarkable (relative to other models). It is slow and expensive, and the "loops are all you need" discourse they are pushing is obvious in the context of someone using Fable-class models What I've found so far is that for broad scope design (code architecture) tasks, Fable is unremarkable. Or, not better enough to justify its cost and speed. But in highly targeted goal-oriented loops, it is another beast entirely. It is very slow but produces very good results. I let it churn on optimizing a SwiftUI-layout resolver in Go I wrote and it was able to bring it down to an order of magnitude I could not reach myself (micro => nanosecond scale). But it took 2 hours and $40 to do it and I had to claw back some changes it overfit to Apple Silicon. Still, very worth it. In comparison, for "implement this feature/change" iterative work, I ran head-to-head Fable vs GPT5.5 vs. GLM-5.1. They all produced equally acceptable final results, but GPT5/GLM did it in a couple minutes and Fable was churning away for 40 minutes. And GLM cost me less than a dollar, GPT5.5 ~$1.50, and Fable cost $9. You can see that in this context, interactively working with an agent is nonsense. Its too slow. You need to write loops to keep the agent working and you probably want to highly parallelize the work being done. As with all things, I think a balance makes sense... My sense is that I'd reserve Fable for targeted, surgical analysis and work. Not for daily driving everyday tasks. I'm going to keep spending a shitload of money (relatively) and maining Fable for the rest of the week to continue to judge, will report if anything changes. I'll continue to head-to-head as well.

I haven’t had time to dive deeply into all the WWDC changes yet but I’m seeing positive feedback 9 times for every 10 posts. Superficially looks great. Polish on iOS and macOS looks great. Siri looks great. CoreAI and related looks great. Is Apple software so back?

Re: the various "bad VC" stories circling around social media today. I believe it. I have my own stories. I think bigger picture though I'd tell all founders: there's going to be bad shit that you witness. Don't associate. Find the good people because there are plenty. In most cases, you won't be in a position to be able to say anything about the bad things you witnessed. Even @eastdakota chose not to say anything until he was in a position of being CEO of a $80 to $100B market cap company. (To be clear, I'm sure Matthew would've said something earlier, he just didn't care to. But surely at some point in history he was in a position he didn't feel comfortable to as well.) Just move on, find the good people. Bide your time. But most importantly, do not associate with the bad people. Maybe they get away with it maybe they don't, but people definitely KNOW who the bad people are and you don't want to be in those rings. It's not worth it.

✌️

My teeth were clenched, but Tesla FSD just reversed out of my garage through a curved driveway with less than 2 inches of clearance on either side with a brick wall and my wife’s SUV. Crazy work. I knew it’s possible cause I do it regularly. But it’s a lot of work, a lot of adjustment cause you have that 2 inch clearance through a curve. I started today and was like, you know what, I’m tired this morning. I’m going to let Jesus take the wheel. And it did it. So my mornings are about to become significantly more chill. I have a video but don’t want to dox myself so can’t share lol.

If you want to know what giveaway LLM design is, this is it: thin colored borders, gradients, glow effects, too many different font sizes, small fonts too small, inconsistent padding and alignment (especially vertical). Not a dunk on @zeeg he’s being transparent about it. And he’s a good AI driver and good developer in general. But using this moment to show people how obvious this is.

There’s a ton of non-technical criticisms I have about aliens.gov but I generally don’t use my X for political commentary. So I’ll say from a technical side that this website is really ugly and poorly implemented. I found a half dozen bugs in seconds. Embarrassing.

I don't know who these people are or who they represent, but they're sniffing around trying to learn about me behind my back. Lots of people I've worked with have been pinging me. Don't know what they're up to, but if you get one go ahead and tell them I'm fucking awful. Thanks.

I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem. As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)! I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work. It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results? 88ms => 1.5ms 150K allocs => ~500 allocs Incredible right? Nope. My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path. This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput. The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity. Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.

As the family IT guy its so disappointing how bad of an experience technology is for non-technical people. I had the distinct pleasure of building educational software for kids full time for a summer while in college (s/o to @WilliamsonMark), and I remember they did weekly/biweekly user testing where a group of toddlers would come in and we'd record them using the software in various states and then adjust accordingly. Every single session was SHOCKINGLY illuminating. Like, I expected after a number of these I'd empathize more and build better toddler software one-shot right? Hell fucking no. Every user study was so educational. I learned I simply can't enter the mind of a toddler. Do TV companies, Netflix/Roku/etc. do user studies with elderly people? Do they realize how dogshit and impossible to navigate their interfaces are? Asking some elderly family members to "sign up and schedule an Uber to pick you up for the airport" is like mission impossible. I thought they were exaggerating, then I tried the experience and holy shit man. Try cold finding, installing, signing up, and scheduling an Uber on a 5 year old iPhone with max font size. Its insane.

Here's the ATC audio clip of my final landing ever. From humble beginnings in a DA40 to ~600 hours in a Vision Jet at 31,000 feet, I love to fly. Due to my growing personal and professional obligations, my final and best ADM decision as a PIC was to stop. This was my toughest landing. The skies were clear. The wind was straight down the runway. It's my home field I've landed at hundreds of times. But I knew it was my last. You can't hear it in the clip but my voice felt shaky. I'm doing my best to focus on a safe landing but I knew that was the last "cleared to land" readback I'd ever give. The landing was smooth. I parked the jet, locked her up, and gave her one last pat on the nose. We had good times. I'm no longer a pilot. Locking in for what's next. 🫡

The new AI web design giveaway is tasteless use of serif fonts plus italics. I mean, there's about 1000 other giveaways cause its all so very ugly, but that's the initial slap in your face giveaway.

This company literally just sends iMessages as a service.

This is why PR diff speed matters. This isn't a dunk on GitHub specifically, because GitLab, Forgejo, etc. are all equal or worse. But this is the kind of thing that drives me nuts, because this is a core workflow and its slow enough I literally take my hands off the keyboard. Btw, when my mouse jiggles on the left, its because the page is literally skipping frames and I'm instinctively shaking my mouse to see if it'll respond. And on the keyboard input you can literally here me finish typing before a letter even shows up. For someone like me who is an expert at these tools, my brain navigates the tool dramatically faster than it can keep up, and that is not good. The tool should not get in the way.

This is how performant PR review could be. On any forge. Pierre is showing us that the only thing holding that back is a skill issue. Excellent ship here! They’re on fire!